4. Dendrochronology in Historic Buildings

Coralie M Mills & Anne Crone

 

State of current knowledge

Dendrochronology has been applied to Scotland’s built heritage since the 1970s, initially focussed on developing native chronology coverage but increasingly applied to providing precision dating and timber provenancing information, an important aspect in a country reliant on imported timber for many centuries. Most individual studies have been published, and several overview publications provide the way into the literature (for example, Baillie 1977; Baillie 1982; Crone 2008; Crone and Mills 2012, Mills and Crone 2012; Crone et al 2017; Mills et al 2017; Mills 2023). Most Scottish building dendro-dates have also been published as summaries in Vernacular Architecture and entered in the journal’s online dendro-database. The dendrochronological work on Scottish buildings has mostly followed the conventional ring-width methods laid out in the English Heritage Dendrochronology Guidelines (1998), with the adoption of newer Blue Intensity (Section 6 Dendroclimatology) and Stable Isotope (Section 8 Stable Isotopes and Dendrochronology) applications for certain projects in recent years.

Few Scottish historic buildings have seen dendrochronological study in contrast to the huge uptake in England and many other countries. Depending on how one counts the sites, given a few large complex sites have had multiple buildings analysed, around 50 Scottish sites have seen such work and are unevenly distributed across the country.

Bar diagram showing all Scottish heritage site chronologies over the last millennium, including buildings and archaeological sites © Created by ScARF for Dendrochronicle/AOC

Outside of regional research projects and HES properties, work on in situ timbers in historic buildings is rare, and much of the extant data comes from ex situ material, either fittings or timbers in collections or samples rescued from timber disposal. Counting fittings and ex situ examples, then material from some 75 individual Scottish buildings has been dendro-dated. In England, the number of analysed buildings is well over 4000 (VAG 2022) due to the longstanding higher levels of support and investment. Considering the dendro-dated building locations by Scottish Research Framework region (Table 1) reveals some interesting regional differences in the numbers of dendrochronological projects and in how the work has come about. South East Scotland contains the most dendro-dated buildings, reflecting in large part the City of Edinburgh Council’s application of archaeological planning conditions requiring dendrochronology, as well as research projects with a southeast focus.

The number of dendro-dated sites in the Highlands is mainly a consequence of the SCOT2K native pine research project (and with a huge gap across the western Highlands and the far north), while the number in North East Scotland is again largely through regional research projects. Although there are only six sites in the Forth Valley, there are 23 individual chronologies due to the large investment by HES in dendrochronology at multiple buildings at Stirling Castle. The two studies in South West Scotland, undertaken in the 1970s when Mike Baillie of Queens University Belfast developed the first native oak chronologies for Scotland, were on the ex situ Lincluden College choir stall boards and on Castle of Park near Stranraer (Baillie 1982) which now appears more likely to represent Irish timber (David Brown pers comm). No further buildings dendrochronology work has occurred in South West Scotland since then. The two sites in Perth & Kinross were on Castle Menzies through the SCOT2K research project (Mills et al 2017) and on late medieval carved oak panels in the Perth Museum, believed to come from St John’s Kirk Perth (Crone et al 2000). There are no dendrochronological dates from buildings in either Argyll & Bute or the SIRFA region (Western Isles, Orkney & Shetland).

 

ScARF Region Number of dendro-dated buildings
South East Scotland 22
Highland 16
North East Scotland 13
Fife 9
Forth Valley 6
Clyde Valley 5
South West Scotland 2
Perth & Kinross 2
Argyll & Bute 0
SIRFA (Scotland’s Islands) 0
Table 1. Numbers of buildings with dendrochronological dates, by ScARF region 

These points are made to highlight the fact that in most regions of Scotland the impacts of development, alteration or repairs on historic timberwork in buildings are not the subject of any archaeological planning condition setting, and the nation is losing much of its timber heritage without record or analysis. There is so much to be learned about individual building histories from expanding the opportunities to dendro-analyse such material, as well as allowing each site to contribute to wider narratives, for example in cultural heritage, building conservation, climatology, forestry and environmental spheres in Scotland. It is still the case that it is easier to identify and date imported timber than native Scottish timber, because of the limited funding opportunities to develop chronologies in Scotland while the countries from which timber was imported historically have seen much better investment for chronology development. Not only does this lack of opportunity inhibit dating here, but it restricts the opportunity to understand the form and fate of Scotland’s lost native woodland cover, a very salient issue when reforestation in Scotland is at the heart of responses to tackling impacts of climate change and biodiversity loss. Ecological restoration could be so meaningfully informed by a fuller understanding of the history and dynamics of our past woodland cover, whilst the tree-ring data themselves can also contribute to reconstructing past climate and long-term woodland ecosystem dynamics.

Location of dendro-dated sites in the last millennium, including buildings and archaeology

An overview of the dendro-buildings work so far (see Crone & Mills 2012; Crone et al 2017; Mills et al 2017; Mills 2023; and recent projects yet to be published) along with later medieval archaeological material reveals the use of local native timber resources in most regions until around the early 15th century, when examples of long distant domestic transport and diverse foreign sources appear.

 

A framework of wooden trusses and roof timbers
St Giles Cathedral, Edinburgh: mid-15th century bell tower frame © Coralie Mills

From the mid-15th century onwards, the numbers of native oak timber structures are few, although probably under-represented given the greater difficulties of dating native material. Eastern and central Scotland increasingly look to Scandinavia for structural oak timber while fine oak boards for carving and painting are brought from the eastern Baltic.

 

A wooden panel carving of a woman holding a child on her knee.
One of the panels from Edzell Castle, Angus. The oak used to make the panels came from the eastern Baltic and was felled sometime after AD 1517. The red line indicates the position of the cast taken off the ring-pattern and the direction of growth is also indicated in red © AOC Archaeology Group

Increasingly through the 16th century and more emphatically in the 17th century, there is a shift to Scandinavian pine, from Norway especially, and then from the mid-18th century to more diverse sources including eastern Baltic pine and American oak. However, this story does not fit all regions, and is dominated by the results from sites in the east, especially around the Forth estuary. Based on limited data, one can so far only vaguely discern the probable longer survival of native timber in some regions like southwest Scotland. Some regional differences are starting to emerge, for example in the character and species of timber being used in buildings, in the timing of the demise of native timber resources, in the sources of imported timber and how those change over time. There is so much more diversity that would be revealed with studies of more buildings, especially in those regions which have so far seen little to no work.

 

Image taken obliqely towards a vaulted timber ceiling in a long narrow room with very high ceiling. The timber planks create a complex arch on the ceiling, supported by timber brackets placed all the way along the wall.
Edinburgh Castle Great Hall roof, built with oak from S Norway felled AD 1509/10 © HES

Unsurprisingly perhaps, the Scottish city with the most dendro-dated buildings is Edinburgh where the results from 12 sites tell a remarkably uniform story. Except for the 15th century dates from St Giles all those buildings that lie along the Royal Mile, from the Castle to Holyrood Palace span only a century, from AD 1509 to 1605, presumably reflecting the population explosion in the capital and the increased need for accommodation during this period. Some of the later dates may represent rebuilding in the aftermath of the Marian riots of 1571-2. The source of the timber in this period was invariably Norway, and it was mostly oak, although pine also occurs. Many of the assemblages display multiple felling dates over a sequential number of years which generally indicates the use of stockpiled timber and reflects the volume of timber being imported from multiple sources into the ports during this period.

 

The underside of a wooden ceiling which is painted with floral and foliate patterns in an orange and cream palette
The painted board-and-beam ceiling over the 2nd floor in 302 Lawnmarket, Edinburgh. The beams are Norwegian pine, felled in AD 1591 and installed some time after that © AOC Archaeology Group

The future potential for exciting new results is indicated by recent unexpected discoveries, such as the 1450s dates and native northeast Scottish origin for the medieval oak bell frame timbers at St Giles in Edinburgh (Mills 2022). The SESOD project (see Case Study: SESOD Project) revealed this to be the earliest dendro-dated building in Edinburgh’s Old Town and, so far, the only extant native timber structure there, the Old Town otherwise characterised by imported timbers in 16th century and later buildings. Even earlier native oak timbers, re-used within Kirkliston Parish Church, on the outskirts of Edinburgh, represent the earliest of any Scottish building dendro-dates, at AD 1203, interpreted as dating the construction of the Norman Transitional church at Kirkliston. Together with St Mary’s in Haddington, where large structural oak was imported from the Gdansk region in the early 15th century, and recycled in later nave roof repairs, these examples from the SESOD project (Mills 2022) show the unexplored promise of hundreds of pre-Reformation Scottish churches. The search for native oak within southeast Scotland for SESOD also showed just how rare it is to find any surviving oak structures, irrespective of source, in this region outside of Edinburgh’s Old Town. This emphasises the value of changing heritage protection systems and finding resources to study historic timbers that are being removed from Scottish historic buildings.

Outside of Edinburgh those buildings that have been dated are more commonly Royal palaces (ie Stirling and Falkland) or lordly residences (ie the castles of Crathes, Darnaway, Drum, Duntarvie, Neidpath, Newark, Rosslyn, Tantallon, Threave, Fenton Tower, Castle of Park, Castle Menzies). Within these buildings it is usually the roof timbers and floor joists that have been dated although fixtures and fittings such as doors and carved panels have been examined at Craigievar, Edzell, Falkland and Stirling, revealing the preference for Baltic oak boards for these purposes. Very few middle-ranking or smaller domestic buildings have been dated; Bay Horse Inn, Dysart is a rare example from the 16th century. From the 17th century onwards, there is a wider range of dated building types, such as Cowane’s Hospital in Stirling, Provan Hall in Easterhouse and functional buildings like the granaries at Foulis and Elie. From the 18th century onwards, there are more examples of dated vernacular buildings. The latter are mostly native pine buildings in the eastern Highlands and upland Aberdeenshire, studied through the SCOT2K project (see Case Study: SCOT2K Project). They include a cluster of pine crucks built in the late 18th to early 19th centuries (Mills et al 2017).

 

A single story derelict cottage with painted white stone walls and a partially exposed wooden roof
Badden Cottage, native pine cruck cottage near Kincraig in 2010 © Coralie Mills

By analysing conifers as well as oak, it becomes apparent how pine (Pinus sylvestris) and sometimes Norway spruce (Picea abies) came to replace oak in the imported timber supplies, especially in the southeast and around the east coast, whilst inland in the Highlands native pine was used from at least the 16th to the 19th centuries, for anything from castle roofs to tiny cruck cottages (Mills et al 2017).

 

The inside of a roof cavity showing triangular wooden frame covered in wooden planks
Badden Cottage native pine roof truss and cabers © Coralie Mills

The dating of native pine structures, often with short sequences, was assisted by the application of Blue Intensity methodology alongside conventional ring-width dendrochronology. In a recent study for HES on two garrison-era buildings at Stirling Castle, pine dendrochronology was combined with detailed analysis of woodworking and shipping marks, the literature suggesting a likely Eastern Baltic origin for the marks; however, dendrochronology showed that the timber was shipped from Norway and provided precise felling dates in the late 18th and early 19th century (Darrah & Mills 2023). This example is mentioned to show that combination of dendrochronology and detailed recording of timber marks and other woodworking evidence can be very powerful in understanding the history, construction and significance of a timber structure, and could be an exemplar for future studies.

 

 

Leave a Reply